What is semantic recording?

Semantic recording is a way of capturing a user session by recording the meaning of the interface, such as each element's type, text and position, rather than a picture of it.

Why it matters

Screenshot-based mobile recording sends a stream of images. Images are heavy, they capture whatever was on screen including secrets, and nobody can search or translate them. Semantic recording keeps the structure instead: a button labelled Continue at these coordinates, a list with these rows.

The result looks like a wireframe with real words. That is enough to see what a user did, and it is text, so input masking can happen on the device and session translation can rewrite the labels for a reviewer in another language.

How AnyReplay handles it

The AnyReplay React Native SDK records the committed React element tree rather than pixels. On each tick it walks that tree, notes what each host element is and what it says, then measures where it sits on screen. No native module, no rebuild, and it works in Expo Go on SDK 52+. Limits are honest: a FlatList records only rendered rows, maps and videos appear as labelled boxes, and custom native components show as Unknown boxes. See session replay for React Native.